Since 1988 the NIMH Genetics Initiative has supported a national resource for the study of bipolar disorder (BP). By 1997 153 multiplex families were assessed, providing cell lines, DNA, and anonymized clinical data. This is now a publicly available resource and analytic results have been published. A second effort commenced in 1998 to ascertain 500 new BP sib pairs and this goal has been exceeded with 523 additional BPI sib pairs ascertained, interviewed, and a DNA sample collected. A genome wide scan has been completed at the Center for Inherited Disease Research (CIDR) on 237 sib pair families and the remaining 309 families will be genotyped by CIDR during 2003. This resource, the largest of its kind, has revealed evidence for areas of linkage on chromosomes 6q and 17q. It has also provided confirmation of a locus on chromosome 22q and support for areas on 1p, 10p, 16p, 13q, and 21q. Accumulating linkage data has implicated other chromosomal regions. We propose an extension of the national genetic resource to include a sample of 5000 unrelated BP probands and 2000 parents for case-control, and family-based association studies. Control samples will be obtained through the NIMH Genetics Initiative national resource. Probands and parents will be ascertained and assessed at eleven sites (the ten sites previously participating plus Howard University, which will provide African-American probands). This sample will be a national resource for fine scale linkage disequilibrium mapping within regions of linkage, as well as candidate gene association studies. Parental DNAs in a subsample will allow control for ethnic stratification. Bioinformatics techniques will be developed and supported for genomic analysis of candidate regions, to assist selection of SNPs and other polymorphic markers (including surrounding and within candidate genes), and primer design. The genotyping will be coordinated across 8 labs with an informed step-wise approach, beginning with standard microsatellite mapping of the current set of 699 pedigrees, followed by contract genotyping of SNPs in an industrial laboratory, and continuing with follow-up genotyping and sequencing of candidate genes and regions in laboratories at the individual sites. SNP typing of the larger case-control sample will occur in the final year of the collaborative study. Analysis of the existing sib pair families plus this large set of cases and controls should permit the confirmation of several vulnerability genes during this grant period.